Methodology — National Benchmark of U.S. Mothers

How we conduct the Benchmark.

Wave 1 — Spring 2026
Download the full methodology (PDF)

About This Page

The National Benchmark of U.S. Mothers is a semiannual, nationally representative research infrastructure tracking five composite indices of maternal and family wellbeing. This page documents the complete methodology for Wave 1. It is published before data collection begins and updated with exact fieldwork dates after the survey closes.

Count on Mothers publishes full methodology with every wave — not as a compliance exercise, but because the integrity of independent research depends on it.

Sample Sourcing & Recruitment

Wave 1 draws from two distinct panels, combined into a single weighted dataset. A third-party survey sampling platform provides access to a representative panel of respondents. The Count panel enriches the dataset with high-engagement respondents and supports longitudinal insight.

Community Panel — The Count

The Count is Count on Mothers' subscriber community — mothers across all 50 states who joined countonmothers.org and have opted in to participate in surveys. Wave 1 invitations are issued to The Count members via direct email and through CoM's Anchor Mom Advisor network.

Recruitment method

Direct email invitation to The Count subscriber list. Personalized invitation to Anchor Mom Advisors (a structured advisory segment of engaged community members representing all U.S. regions and the political spectrum).

Respondent profile

Self-selected from CoM's community; higher engagement and authentic motivation than cold opt-in panels; stronger longitudinal retention.

Primary role in Wave 1

Community panel contributes qualitative depth, enrichment data, and community-sourced insight. Community panel responses are incorporated into the weighted dataset (see Hybrid Design below).

Panel integrity advantage

As AI-generated survey responses threaten paid panel quality industry-wide, Count on Mothers' community-sourced panel is more defensible. Community members with established relationships are more resistant to synthetic responses due to established participant relationships and longitudinal validation signals; longitudinal behavioral patterns enable anomaly detection.
Paid Panel — Third-party Survey Sampling Platform
We use a national online survey panel provider used to recruit a random, nationally representative sample of U.S. mothers for Wave 1. Paid panel respondents are recruited independently of CoM's community and are not self-selected into a CoM relationship.

Recruitment method

Vendor-sourced online panel. Respondents are screened at intake to confirm U.S. residency and current parenting status (mother, child in household).

Compensation

Paid panel respondents receive standard panel incentive.

Role in Wave 1

Primary source for nationally representative quantitative wave. Provides the statistically defensible random sample required for institutional-grade benchmark legitimacy.

Target n

Approximately 2,000–2,500 completed responses from paid panel, sufficient for national topline reporting and key subgroup analysis.

Hybrid Design

Both panels contribute to a single Wave 1 dataset. Separate post-stratification weights are applied to each panel before combination. This approach is methodologically defensible and preserves both the statistical rigor of the paid panel random sample and the community value of The Count panel. All reported topline statistics and index scores are drawn from the combined, weighted dataset.

Weighting Variables, Census Vintage & Method

Weighting Method

Post-stratification raking (iterative proportional fitting) is used to align the combined survey sample with U.S. population targets. Raking iteratively adjusts respondent weights across multiple demographic variables simultaneously until the sample distribution converges with Census benchmarks. This is the standard approach used by major survey organizations including Pew Research Center, Gallup, and NORC.

Census Vintage

Population targets are drawn from the American Community Survey (ACS) 5-Year Estimates, most recent available year at time of weighting. [ACS vintage year to be confirmed and documented here after final weighting is applied.]

Weighting Variables

The following variables are used as weighting targets. All targets are specific to U.S. mothers (not the general population).

Variable

Target Population

Age
Distribution of mothers by age cohort (18–29, 30–39, 40–49, 50+)
Race / Ethnicity
Non-Hispanic White, Non-Hispanic Black, Hispanic/Latina, Asian/Pacific Islander, Other/Multiracial
Census Region
Northeast, Midwest, South, West (4-region; 9-division available for subscriber crosstabs)
Education Level
Less than high school, High school diploma/GED, Some college, Bachelor's degree, Graduate degree
Household Income
Used as supplemental weight where response rate permits; see SES methodology note below
Weight capping: Sensitivity checks are conducted to ensure weight capping does not materially distort population alignment.

Socioeconomic Status Methodology Note

Why We Use Education + Area Deprivation Index (ADI) as Our SES Measure

Count on Mothers uses a combination of education level and zip code–linked Area Deprivation Index (ADI) as our socioeconomic status measure, in addition to self-reported household income.

Self-reported income has well-documented limitations in survey research: non-response rates of 15–25% are typical, and wave-to-wave volatility introduces noise that undermines longitudinal comparability. Our approach — pairing education (stable, well-reported, and strongly predictive of economic conditions) with neighborhood-level deprivation data (ADI, derived from Census-linked geographic indicators) — produces a more stable and reliable proxy for socioeconomic conditions across waves.

This approach is consistent with contemporary practice in population health research and social survey methodology. It enables more confident subgroup comparisons across waves without the volatility introduced by income non-response.

Margin of Error

All margins of error are reported at the 95% confidence level.

Sample

Margin of Error (±, 95% CI)

Full sample (n = 2,000–2,500)
±2.2–2.5 percentage points
Key subgroups (n ≈ 500)
±4.4 percentage points
Smaller subgroups (n ≈ 200)
±6.9 percentage points
Subgroup reporting threshold: Subgroup percentages are not reported when the unweighted n falls below 50. Results are flagged as "directional only" when unweighted n falls between 50 and 99.

‍Cross-partisan reporting: Political ideology subgroup analysis is a core CoM output. When findings are consistent across ideological groups, results are labeled "consistent across all political groups" — CoM's strongest credibility signal for policy and institutional audiences.

Effective Sample Size

Effective sample size (n_eff) is reported alongside unweighted n for all wave publications and institutional deliverables.

Post-stratification weighting increases the representativeness of the sample but introduces design effects: some respondents receive higher weights than others, which reduces the statistical precision compared to a pure random sample of the same size. The effective sample size accounts for this:
n_eff = n / DEFF
Effective sample size is reported in the methodology appendix of every wave report. Institutional subscribers receive n_eff alongside unweighted n and weighted n in the full data deliverable.

Index Construction

The National Benchmark tracks five composite indices, each scored on a 0–100 scale. For all five indices: higher scores indicate better conditions (lower stress, stronger trust, greater access, safer environment). This directional convention is held constant across all waves.

Each index is constructed from a defined set of fixed survey items. Raw item responses are standardized, directionally aligned, and aggregated into a composite score using weighted averaging. Individual item weights within each index are documented in the methodology appendix and held constant across waves.

The Five Indices

MCSI

Maternal Capacity & Stress Index

What it measures
The composite of maternal capacity, strain, and buffering resources — perceived stress, sleep and time pressure, confidence handling unexpected family challenges, administrative cognitive load, and the social support mothers have to draw on. Captures the operating condition mothers are functioning in, not just their reported stress level.
Why it matters
MCSI is the leading indicator institutional partners use to interpret movement in every other index. When MCSI declines, the significance of pressures captured in FEPI, ITAI, and YECCI amplifies — economic strain, institutional friction, and digital concerns all hit harder when maternal bandwidth is low. For organizations designing services, advocacy strategy, or policy interventions for families, MCSI surfaces the population conditions under which other interventions will succeed or fail.

FEPI

Family Economic Pressure Index

What it measures
Household economic pressure as mothers experience it — across housing, childcare, healthcare, and food, and the care families delay or skip when costs are out of reach. Measures pressure as a persistent condition, not as a function of nominal income, capturing the squeeze families feel across income levels.
Why it matters
Income-based measures consistently undercount the economic pressure mothers actually navigate, because cost burden across childcare, healthcare, and housing concentrates structurally in families with children regardless of income tier. FEPI gives funders, coalitions, and policy organizations a measurement of family economic strain that doesn't disappear when income rises — surfacing the populations who appear "stable" on income measures but are actually carrying significant pressure.

ITAI

Institutional Trust & Accountability Index

What it measures
Maternal trust across institutions and industries — schools, healthcare, health insurance, food and beverage, pharmaceutical and drug manufacturers, news sources, technology platforms, and government — and whether mothers can navigate these institutions effectively to get what their families need. Captures trust as a perception of whether systems are set up to act in the best interests of children and families.
Why it matters
Institutional trust predicts engagement with services, willingness to seek help, and civic participation. For institutions whose work depends on family uptake — health systems, school districts, government benefits programs, advocacy coalitions — ITAI surfaces where the trust gap lives, who carries it most, and whether it moves in response to policy or institutional change. Cross-partisan patterns in ITAI are among the most actionable findings the Benchmark produces.

YECCI

Youth Environment & Commercial Conditions Index

What it measures
Structural conditions shaping children — digital and AI exposure, online platform behaviors, commercial design and youth-directed marketing, neighborhood safety, and generational parenting norms — across the environments mothers navigate for their kids. Captures the maternal vantage on how today's youth environment is functioning.
Why it matters
YECCI gives organizations working in child wellbeing, online safety, family policy, and youth-directed industries a measurement of the environmental conditions affecting children that comes directly from mothers — the audience these institutions are trying to reach, persuade, or serve. The maternal vantage matters because mothers see what isn't visible from outside the home. For companies designing for children, regulators evaluating duty-of-care frameworks, and coalitions advocating for youth protections, YECCI tracks where maternal concern is concentrated and how it shifts.

CWASI

Child Wellbeing Access & Support Index

What it measures
Children's mental, emotional, and behavioral health access — the school and insurance gateways mothers depend on, peer relationships, and the friction families face when seeking help. Captures both whether support is structurally available and whether mothers can actually reach it through the institutions designed to deliver it.
Why it matters
The access gap between needing support for a child and getting it is where most family interventions succeed or fail. For health systems, school districts, foundations, and advocacy organizations working on child mental health, CWASI surfaces where the gap is widest, who carries it most, and which institutional gateways — schools, insurance, providers — are working or failing for families. The data directly informs program design, policy advocacy, and resource targeting.
Complete item-level composition for each index — including individual survey items, response scales, directional coding, and item weights — is documented in the full methodology appendix available to institutional subscribers.

Panel Integrity Protocols

As AI-generated and bot-produced survey responses become a growing threat to online panel quality industry-wide, Count on Mothers treats panel integrity as a competitive differentiator — not just a quality control step. The following protocols are applied to every wave.

Speed Screener

All survey completions are flagged if total response time falls below a minimum threshold calibrated to the instrument length. Wave 1 threshold: completions under 4 minutes are flagged for review.
Flagged responses are individually reviewed and removed from the dataset if they fail additional quality checks.

Open-Ended Response Quality Control

Wave 1 includes at least one open-ended response question. All open-ended responses are screened for: (a) AI-generated or templated language patterns; (b) copy-paste duplication across respondents; (c) off-topic or nonsensical content.
Responses failing this screen are removed from the dataset. The open-ended QC step provides a high-sensitivity signal for detecting synthetic respondents that closed-ended items alone cannot detect.

Attention Check Items

The Wave 1 instrument includes embedded attention check items — instructed-response questions (e.g., "For this question, please select [specific answer]") placed within the survey flow.
Respondents who fail attention checks are flagged. Respondents who fail more than one attention check are removed from the dataset prior to weighting.

Community Panel Longitudinal Integrity

The Count community panel has structural integrity advantages over cold opt-in panels. Community members have established relationships with CoM over time, creating behavioral fingerprints that enable anomaly detection. Longitudinal wave-over-wave participation patterns are monitored for irregular activity. New community panel entrants for Wave 1 are subject to the same speed, open-ended, and attention check protocols as the paid panel respondents.

Why This Matters for Institutional Buyers

The methodological risk of AI-contaminated survey panels is not theoretical — it is actively degrading the quality of opt-in panel research industry-wide. Count on Mothers' blended panel architecture, combined with multi-layer integrity screening, produces a more defensible dataset than single-source cold panels. We document and publish these protocols because institutional buyers purchasing independent data should be able to evaluate what they are buying.

Instrument Stability

Wave 1 establishes the baseline. It is the foundation from which all trend analysis is built.

One wave = report.
Two waves = trend.
Three waves = infrastructure.

The Wave 1 core instrument consists of 25–35 fixed questions. These questions are held constant across waves. The stability of the core instrument is what makes longitudinal comparison possible — and what makes the Benchmark's value compound with each subsequent wave.
  • Fixed core questions: Do not change across waves without a documented rationale, a formal review by the research team and Methodology Advisory Board, and an overlap procedure (running both old and new wording in the transition wave on split samples) to establish trend continuity before retiring any item.
  • Response scales: Held constant within subdomains. Mixing scale types across waves introduces measurement error.
  • Index construction: Item composition and weighting within each index is documented and locked at Wave 1. Changes to index construction require the same overlap procedure as item changes.

Rotating Modules

Wave 1 includes a supplemental rotating module of up to 4 questions (plus an optional open-response field). Rotating module questions address time-sensitive topics and are positioned after the fixed core instrument. They are clearly labeled as supplemental in all published materials.
  • Independence from core scores: Rotating module responses are not incorporated into the five composite index scores. They are analyzed and reported separately.
  • Labeling: All findings from the rotating module are labeled "Supplemental — Wave [n] Rotating Module" in published materials and data deliverables.
  • Institutional add-on: Rotating module topic selection is eligible for institutional underwriting as an à la carte add-on. The CoM research team retains final authority over question wording, framing, and inclusion.

Research Team

Institutional subscribers and data partners are buying trust in the methodology. Research team members' credentials are displayed prominently in every wave report and institutional deliverable.
Jennifer Brailsford, PhD
Director of Research
Survey methodology and research design. Responsible for instrument architecture, question development, index construction, and methodology documentation.
Melissa Lawrence, MPH
Director of Data Science
Statistical weighting, data cleaning, and quantitative analysis. Responsible for raking procedures, effective sample size calculation, and longitudinal dataset architecture.
Academic partnerships supporting Wave 1 methodology:
‍
University College London (Methodology Advisory Board — Dr. Kaitlyn Regehr, Dr. Photini Vrikki, Dr. Katharine Smales).

Fieldwork Dates

Wave 1 Field Dates

Survey open:
[TO BE ADDED AFTER FIELDWORK CLOSES]
Survey close:
[TO BE ADDED AFTER FIELDWORK CLOSES]
Target field window:
14–21 days
This page is published before fielding begins and updated with exact open and close dates after the survey window closes.

Sample Growth & Longitudinal Transparency

The Benchmark is designed to grow in analytical power across waves.
Wave 1 establishes the baseline at n = 2,000–2,500 (±2.2–2.5 pp). As the CoM community panel grows through ongoing subscriber recruitment, subsequent waves will increase sample size — reducing margin of error and enabling more granular subgroup analysis at the regional, demographic, and political levels.
This growth is a designed feature of the Benchmark architecture, not a concession.
The paid panel provides a stable representativeness floor at any sample size. Community panel growth compounds  the dataset’s analytical power without replacing the statistical rigor of the random sample foundation.
n = 2,000–2,500
Wave 1 sample size (±2.2–2.5 pp at 95% CI)
Wave-over-wave sample sizes and methodology changes are documented in a public change log maintained on this page beginning with Wave 1. Pre-Wave 1 instrument iteration history is documented in the institutional methodology appendix and is not trend-affecting. From Wave 1 forward, no change to core instrument questions, index construction, or weighting methodology is made without a formal overlap procedure and published rationale.

Methodology Changes

Wave

Change Date

Change Summary

Rationale

Wave 1
Spring 2026
Baseline established. No prior waves to compare.
—

Questions about our methodology?

Email the research team Join The Count
Count on Mothers • National Benchmark of U.S. Mothers • Wave 1 Methodology • Spring 2026
COUNT ON MOTHERS
National Benchmark of U.S. Mothers
Wave 1 Methodology • Spring 2026